From: Jeff King <peff@peff.net>
Date: Fri, 22 Feb 2013 17:27:10 GMT
Subject: Re: [PATCH] archive: let remote clients get reachable commits
Message-ID: <20130222172710.GB17475@sigill.intra.peff.net>
In-Reply-To: <7vehg8s295.fsf@alter.siamese.dyndns.org>

On Fri, Feb 22, 2013 at 09:10:46AM -0800, Junio C Hamano wrote:

> I personally think that it is OK to limit the scope to expressions
> that start from the tip of ref and expressions that start with the
> SHA-1 at the tip of ref, e.g.
> 
>     master~12:Documentation
>     v2.6.11:arch/alpha
>     5dc01c595e6c6ec9ccda	# tag v2.6.11
>     26791a8bcf0e6d33f43a:arch	# tag v2.6.12
>     26791a8bcf0~12:arch		# starting at 26791a8b and dig down
> 
> are OK, while forbidding the following:
> 
>     c39ae07f393806ccf406        # tree of tag v2.6.11
>     9ee1c939d1cb936b1f98	# commit v2.6.12^0
>     9ee1c939d1cb936b1f98:	# tree of commit v2.6.12^0
>     9ee1c939d1cb936b1f98:arch	# subtree of commit v2.6.12^0
> 
> which will make it significantly easier to implement the necessary
> validation in a robust way.

How are you proposing to verify master~12 in that example? Because
during parsing, it starts with "master", and we remember that? Or
because you are doing a reachability traversal on all refs after we
parse?

-Peff